*** Reload Knowledge Graph *** Deserializing compressed entity integer mapping... Done! Deserializing compressed relation integer mapping... Done! Converting integer and relation mappings from from pandas dataframe to dictionaries for an easy access... Done! Deserializing integer mapped data and mapping it to numpy ndarray... Done! Deserializing integer mapped data and mapping it to numpy ndarray... No valid data found Deserializing integer mapped data and mapping it to numpy ndarray... No test data found /home/c/cdemir/profiles/unix/cs/CD/DAIKIRI-Embedding/DAIKIRI_Storage/2022-03-09 13:41:36.868332 Preprocessing took: 184.286 seconds ------------------- Description of Dataset None ------------------- Number of entities: 114747963 Number of relations: 13906 Number of triples on train set: 375900264 Number of triples on valid set: 0 Number of triples on test set: 0 ------------------- Train & Eval ------------------- Loading pre-trained model... Training starts: QMult-labeling:NegativeSampling Creating training data... Number of mini-batches to compute for a single epoch: 752 QMult( (loss): BCEWithLogitsLoss() (emb_ent_real): Embedding(114747963, 25) (emb_ent_i): Embedding(114747963, 25) (emb_ent_j): Embedding(114747963, 25) (emb_ent_k): Embedding(114747963, 25) (emb_rel_real): Embedding(13906, 25) (emb_rel_i): Embedding(13906, 25) (emb_rel_j): Embedding(13906, 25) (emb_rel_k): Embedding(13906, 25) (input_dp_ent_real): Dropout(p=0.0, inplace=False) (input_dp_ent_i): Dropout(p=0.0, inplace=False) (input_dp_ent_j): Dropout(p=0.0, inplace=False) (input_dp_ent_k): Dropout(p=0.0, inplace=False) (input_dp_rel_real): Dropout(p=0.0, inplace=False) (input_dp_rel_i): Dropout(p=0.0, inplace=False) (input_dp_rel_j): Dropout(p=0.0, inplace=False) (input_dp_rel_k): Dropout(p=0.0, inplace=False) (hidden_dp_real): Dropout(p=0.0, inplace=False) (hidden_dp_i): Dropout(p=0.0, inplace=False) (hidden_dp_j): Dropout(p=0.0, inplace=False) (hidden_dp_k): Dropout(p=0.0, inplace=False) (bn_ent_real): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_ent_i): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_ent_j): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_ent_k): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_rel_real): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_rel_i): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_rel_j): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_rel_k): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_hidden_real): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_hidden_i): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_hidden_j): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (bn_hidden_k): BatchNorm1d(25, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) | Name | Type | Params --------------------------------------------------------- 0 | loss | BCEWithLogitsLoss | 0 1 | emb_ent_real | Embedding | 2.9 B 2 | emb_ent_i | Embedding | 2.9 B 3 | emb_ent_j | Embedding | 2.9 B 4 | emb_ent_k | Embedding | 2.9 B 5 | emb_rel_real | Embedding | 347 K 6 | emb_rel_i | Embedding | 347 K 7 | emb_rel_j | Embedding | 347 K 8 | emb_rel_k | Embedding | 347 K 9 | input_dp_ent_real | Dropout | 0 10 | input_dp_ent_i | Dropout | 0 11 | input_dp_ent_j | Dropout | 0 12 | input_dp_ent_k | Dropout | 0 13 | input_dp_rel_real | Dropout | 0 14 | input_dp_rel_i | Dropout | 0 15 | input_dp_rel_j | Dropout | 0 16 | input_dp_rel_k | Dropout | 0 17 | hidden_dp_real | Dropout | 0 18 | hidden_dp_i | Dropout | 0 19 | hidden_dp_j | Dropout | 0 20 | hidden_dp_k | Dropout | 0 21 | bn_ent_real | BatchNorm1d | 50 22 | bn_ent_i | BatchNorm1d | 50 23 | bn_ent_j | BatchNorm1d | 50 24 | bn_ent_k | BatchNorm1d | 50 25 | bn_rel_real | BatchNorm1d | 50 26 | bn_rel_i | BatchNorm1d | 50 27 | bn_rel_j | BatchNorm1d | 50 28 | bn_rel_k | BatchNorm1d | 50 29 | bn_hidden_real | BatchNorm1d | 50 30 | bn_hidden_i | BatchNorm1d | 50 31 | bn_hidden_j | BatchNorm1d | 50 32 | bn_hidden_k | BatchNorm1d | 50 --------------------------------------------------------- 11.5 B Trainable params 0 Non-trainable params 11.5 B Total params 45,904.750Total estimated model params size (MB) [1 / 1] Training is started.. Validation sanity check: 0it [00:00, ?it/s] Training: 0it [00:00, ?it/s] Training: 0%| | 0/752 [00:00